GPT-5.1-Codex
GPT-5.1-Codex is a GPT-5.1 generation coding agent model designed for autonomous software engineering, combining improved reasoning over the GPT-5 codex generation with the ability to read, write, execute, and verify code in sandboxed environments.
import { streamText } from 'ai'
const result = streamText({ model: 'openai/gpt-5.1-codex', prompt: 'Why is the sky blue?'})Playground
Try out GPT-5.1-Codex by OpenAI. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by OpenAI
| Model |
|---|
| 1M | 3.2s | 68tps | $5.00/M | $30.00/M | Read: $0.5/M Write: — | $10.00/K + input costs | — |   | 04/24/2026 | ||||
| 400K | 1.5s | 190tps | $0.75/M | $4.50/M | Read:$0.07/M Write:— | $10.00/K + input costs | — | | 03/17/2026 | ||||
| 400K | 0.5s | 117tps | $0.20/M | $1.25/M | Read:$0.02/M Write:— | $10.00/K + input costs | — | | 03/17/2026 | ||||
| 128K | 0.5s | 111tps | $1.25/M | $10.00/M | Read:$0.13/M Write:— | $10.00/K + input costs | — | | 11/12/2025 | ||||
| 400K | 3.4s | 446tps | $0.25/M | $2.00/M | Read:$0.03/M Write:— | $14/K + input costs | — | | 08/07/2025 | ||||
| 131K | 0.1s | 223tps | $0.35/M | $0.75/M | Read:$0.25/M Write:— | — | — |    +5 | 08/05/2025 |
About GPT-5.1-Codex
GPT-5.1-Codex was released on November 12, 2025 as the standard tier in OpenAI's GPT-5.1 codex family on AI Gateway. It sits between the more affordable codex mini and the maximum-compute codex max, giving teams a middle-ground option that balances capability with cost.
The GPT-5.1 generation brought deeper reasoning and more reliable instruction following to the codex architecture. For GPT-5.1-Codex, that translates to better understanding of complex codebases, stronger adherence to existing coding conventions, and more accurate multi-step problem solving. The model operates in the agentic coding loop: reading repository context, planning changes, writing code, executing tests in sandboxed environments, and iterating.
With a context window of 400K tokens, GPT-5.1-Codex can process substantial repository contexts in a single pass, enabling it to understand broader architectural implications rather than operating on isolated files.
What To Consider When Choosing a Provider
- Configuration: GPT-5.1-Codex builds on the GPT-5 codex foundation with the GPT-5.1 generation's improvements in reasoning and instruction following, translating to better code quality and more reliable autonomous task completion.
- Configuration: For maximum compute on the hardest coding problems, consider GPT-5.1 codex max. For routine tasks, codex mini provides faster, cheaper results.
- Zero Data Retention: AI Gateway supports Zero Data Retention for this model via direct gateway requests (BYOK is not included). To configure this, check the documentation.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use GPT-5.1-Codex
Best For
- Autonomous feature development: Writing complete features from specifications with test verification
- Complex debugging: Tracing issues through large codebases and producing verified patches
- Multi-file refactoring: Changes that span modules and require understanding architectural patterns
- Code migration: Converting codebases between frameworks, languages, or API versions
- Comprehensive test generation: Creating thorough test suites that cover edge cases and failure modes
Consider Alternatives When
- Hardest coding challenges: GPT-5.1 codex max applies maximum compute for the most demanding tasks
- Routine coding tasks: Codex mini handles simpler bug fixes and scaffolding more efficiently
- General-purpose work: GPT-5.1 instant or thinking for non-coding tasks
- Budget constraints: Earlier codex models at lower price points for cost-sensitive pipelines
Conclusion
For teams that want GPT-5.1 generation coding capability without the compute premium of codex max, GPT-5.1-Codex is the practical middle ground. It handles autonomous coding workflows reliably and slots into AI Gateway alongside other codex variants.
Frequently Asked Questions
How does GPT-5.1-Codex improve over GPT-5 codex?
It benefits from the GPT-5.1 generation's advances in reasoning and instruction following, producing better code quality and more reliable autonomous task completion.
When should I use codex max instead?
When tackling the most complex coding challenges where maximum compute and reasoning depth are worth the additional cost, such as large-scale architectural changes or critical security audits.
What context window does GPT-5.1-Codex support?
400K tokens, enabling comprehensive codebase understanding in a single pass.
Can GPT-5.1-Codex run tests?
Yes. It operates in sandboxed environments where it can execute code and run test suites to verify its output.
How does AI Gateway handle authentication for GPT-5.1-Codex?
AI Gateway accepts a single API key or OIDC token for all requests. You don't embed OpenAI credentials in your application; AI Gateway routes and authenticates on your behalf.
What are typical latency characteristics?
This page shows live throughput and time-to-first-token metrics measured across real AI Gateway traffic.